This groundbreaking study tracked the performance of three versions of the Stable Diffusion model, which is widely used in generative AI image generation. Researchers established an “imitation threshold” using an algorithm designed to discern whether a generated image could convincingly be identified as an imitation. This threshold was then corroborated through human evaluation, showing a notable alignment between computer assessments and human judgments.

The research, titled “How Many Van Goghs Does It Take to Van Gogh? Finding the Imitation Threshold,” was published on the preprint server arXiv last month by a team led by Sahil Verma, a computer science PhD student at the University of Washington. Verma commented on the surprising nature of the findings, explaining, “Some people are surprised that it’s such a low number, and some people are also surprised that it’s a high number.”

Particularly fascinating is the detail that replicating the distinct brushstrokes of iconic artists like Vincent Van Gogh can be achieved with as few as 112 images, while producing human likenesses can be done with only 234 images. This efficiency raises serious considerations regarding the implications for copyright infringement as AI technologies become more powerful and ubiquitous.

As the landscape of digital art and its legal underpinnings shifts, the study reflects a growing concern within the art community. Verma noted, “As we were working on it [the study], we realized this has a huge implication for privacy and copyright issues.” This revelation comes at a time when there has been a surge in copyright litigation concerning AI-generated content in the U.S., emphasizing the urgent need to navigate the intersection of technology and intellectual property rights.

In August, a notable ruling allowed a group of artists to proceed with their copyright claims against several AI image generators, including Stability AI, the creator behind Stable Diffusion. The lawsuit, filed by ten artists, claims these companies used their works to develop AI art tools without obtaining the necessary permissions, highlighting a critical area of contention as the technology evolves.

As generative AI continues to push the boundaries of creativity and art replication, the stakes for artists and their rights are set to rise. With the advent of such sophisticated techniques revealing that just a handful of images may suffice for imitation, the conversation surrounding copyrights and AI’s role in the creative landscape is more urgent than ever.

The topic of editing pre-existing images with artificial intelligence has gained significant traction recently, prompting vital discussions about the ethical implications and potential consequences of such technology. Major platforms, including Meta, have been grappling with the challenge of effectively labeling images altered with AI tools versus those generated entirely by AI models. Similarly, Google has rolled out robust AI features without providing clear visual indicators that the images have undergone modifications.

In a notable commitment last year, Midjourney adopted the IPTC’s Digital Source Type property—a technical standard for embedding metadata in images that indicates whether they have been AI-generated. Despite this, Midjourney remains one of the few leading AI platforms that has not yet integrated C2PA, a metadata framework designed to trace an image’s complete provenance, documenting the equipment and software involved in its creation.

On Midjourney’s official Discord channel, CEO David Holtz shared insights about the new image editing feature, revealing that access will initially be restricted to a “subset of the current community.” The announcement highlighted the introduction of increased human moderation alongside new, more advanced AI moderators aimed at curbing potential misuse. Though Holtz candidly admitted uncertainty in precisely restricting this feature’s deployment, he encouraged community feedback through a poll to identify which users would be granted early access.

However, there are significant risks associated with launching these editing tools without adequate oversight. The potential for widespread copyright infringement and the rise of misleading deepfakes form the crux of growing concerns. Deepfakes have proliferated on social media, making it increasingly challenging to differentiate between fact and fabricated content. Following Hurricane Helene, a tidal wave of fake AI-generated images depicting destruction surged across the web.

Data from Clarity, a deepfake detection firm, indicates a staggering 900% increase in the creation and publication of deepfakes this year compared to the same period last year, revealing the alarming scale of the issue. A recent YouGov poll highlighted that 85% of Americans expressed apprehension regarding the spread of misleading deepfakes online.

While the U.S. currently lacks a federal law against deepfakes, over ten states have implemented laws targeting AI-facilitated impersonation. Notably, California is working on legislation that would allow judges to compel the removal of deepfakes from platforms, potentially imposing monetary penalties on offenders.

Despite these rising concerns, Midjourney’s track record has drawn scrutiny amid accusations regarding its alleged use of copyrighted content for training its generative AI models. Nevertheless, the platform has reportedly been taking steps in recent months to mitigate the spread of deepfakes, including enforcing filters on images of political figures, especially ahead of the upcoming U.S. presidential election.

With its upcoming tools set to fuel further discussions on image authenticity and copyrighting in the digital age, Midjourney is at the forefront of a rapidly evolving landscape where the intersection of AI and image manipulation brings both exciting possibilities and pressing challenges.